legacy_scholarly v3 Evaluation

Generated: 2026-09-20 11:22:53 UTC

Pairs100
Distinct lemmas100
Slope0.926
R^20.986
BLEU-452.9%
chrF++73.4%
METEOR76.6%
ROUGE-L78.2%
BERTScore96.2%
COMET79.8%
BLEURT76.2%

Profile version id: 2130

First translation using this version: 2026-05-07

Total usable AI runs for this version: 381 runs across 380 lemmas

Created: 2026-05-07 16:52:06.734553+10:00

Active: True

Notes: Reviewed house-style prompt v3 revised after Gabe's 2026-04-30 feedback and April 30 group discussion. Keeps static rules for global house style, citation hygiene, transliteration, metalinguistic precision, and syntax; moves most formula/gloss/geographic vocabulary decisions to entry-specific guidance context; adds the settlement-location "in" convention, stricter Strabo/modern-locator handling, and revised rules for prose/verse quotation formatting and inflected or dialectal evidence.

Prompt length scatter plot

Length Regression

FormulaAI words = intercept + slope * human words
Slope0.926
Intercept1.985
R^20.986
P-value4.18e-93
Pearson r0.993
Mean source words30.400
Length fallback rows0
Mean human words44.150
Mean AI words42.860
Mean absolute length residual2.760
Mean absolute percent error5.9%

Metric vs Passage Length

These plots test whether score changes with source passage length for this prompt version. The dashed line is an ordinary least-squares fit; the annotation reports n, r, R^2, p-value, and slope.

BLEU-4 vs passage length
BLEU-4: significant negative; r = -0.366, R^2 = 0.134, p = 0.0002.
chrF++ vs passage length
chrF++: significant negative; r = -0.392, R^2 = 0.153, p = 5.61e-05.
METEOR vs passage length
METEOR: significant negative; r = -0.399, R^2 = 0.159, p = 3.99e-05.
ROUGE-L vs passage length
ROUGE-L: significant negative; r = -0.419, R^2 = 0.176, p = 1.42e-05.
BERTScore vs passage length
BERTScore: significant negative; r = -0.576, R^2 = 0.332, p = 3.57e-10.
COMET vs passage length
COMET: significant negative; r = -0.381, R^2 = 0.145, p = 9.03e-05.
BLEURT vs passage length
BLEURT: significant negative; r = -0.703, R^2 = 0.494, p = 3.56e-16.
Trigram precision vs passage length
Trigram precision: significant negative; r = -0.278, R^2 = 0.077, p = 0.0051.
Trigram recall vs passage length
Trigram recall: significant negative; r = -0.284, R^2 = 0.081, p = 0.0041.
Trigram F1 vs passage length
Trigram F1: significant negative; r = -0.282, R^2 = 0.080, p = 0.0045.
Trigram Jaccard vs passage length
Trigram Jaccard: significant negative; r = -0.277, R^2 = 0.077, p = 0.0053.
Metric Rows Pattern Pearson r R^2 P-value Slope Status
BLEU-4 100 significant negative -0.366 0.134 0.0002 -0.00269 ok
chrF++ 100 significant negative -0.392 0.153 5.61e-05 -0.00174 ok
METEOR 100 significant negative -0.399 0.159 3.99e-05 -0.00185 ok
ROUGE-L 100 significant negative -0.419 0.176 1.42e-05 -0.00175 ok
BERTScore 100 significant negative -0.576 0.332 3.57e-10 -0.00041 ok
COMET 100 significant negative -0.381 0.145 9.03e-05 -0.00073 ok
BLEURT 100 significant negative -0.703 0.494 3.56e-16 -0.00266 ok
Trigram precision 100 significant negative -0.278 0.077 0.0051 -0.00217 ok
Trigram recall 100 significant negative -0.284 0.081 0.0041 -0.00220 ok
Trigram F1 100 significant negative -0.282 0.080 0.0045 -0.00218 ok
Trigram Jaccard 100 significant negative -0.277 0.077 0.0053 -0.00219 ok

Translation Similarity Metrics

Mean BLEU-452.9%
Mean chrF++73.4%
Mean METEOR76.6%
Mean ROUGE-L78.2%
Mean BERTScore96.2%
Mean COMET79.8%
Mean BLEURT76.2%
Legacy smoothed corpus BLEU48.9%
Unigram F178.4%
Bigram F155.8%
Trigram precision42.7%
Trigram recall41.4%
Trigram F142.1%
Trigram Jaccard26.6%
4-gram F133.1%

Largest Length Residuals

Lemma ID Headword Human words AI words Residual BLEU-4 chrF++ METEOR ROUGE-L BERTScore COMET BLEURT Human Run Model
2484 Καρία 270 231 -20.959 35.7% 61.7% 64.6% 68.3% 92.0% 80.8% 51.6% reviewed/approved 7782 gpt-5.5
2604 Καρχηδών 117 126 15.693 32.8% 60.9% 72.1% 66.9% 93.1% 79.7% 66.7% reviewed/approved 7787 gpt-5.5
2603 Κάρυστος 175 178 13.995 41.7% 68.3% 71.1% 71.4% 95.2% 79.3% 57.7% reviewed/approved 7786 gpt-5.5
2465 Κάνωπος 83 89 10.171 57.7% 77.4% 87.8% 83.7% 95.8% 78.6% 74.9% reviewed/approved 7488 gpt-5.5
2056 Καβειρία 112 97 -8.678 34.2% 69.5% 55.9% 67.0% 95.7% 79.1% 61.7% reviewed/approved 7761 gpt-5.5
2328 Καλλίπολις 43 50 8.205 38.3% 65.3% 71.2% 66.7% 94.5% 75.4% 71.7% reviewed/approved 7460 gpt-5.5
2605 Κάσιον 56 62 8.169 43.3% 67.3% 72.7% 69.5% 94.1% 74.3% 70.9% reviewed/approved 7788 gpt-5.5
2346 Κάναι 77 81 7.726 34.6% 55.7% 53.0% 63.3% 92.2% 71.9% 50.1% reviewed/approved 7477 gpt-5.5
2470 Καππαδοκία 68 72 7.059 18.4% 55.5% 76.2% 75.7% 93.6% 74.3% 62.5% reviewed/approved 7492 gpt-5.5
2609 Κάσπειρος 135 134 7.028 40.1% 64.0% 57.8% 63.9% 91.1% 73.3% 61.2% reviewed/approved 7790 gpt-5.5
2468 Καπετώλιον 138 123 -6.749 18.1% 45.0% 42.2% 50.6% 90.8% 73.3% 55.9% reviewed/approved 7781 gpt-5.5
7251 Κύρτος 66 57 -6.089 46.5% 65.5% 69.7% 74.8% 94.0% 74.3% 65.1% reviewed/approved 7816 gpt-5.5
2342 Κάμιρος 54 58 6.021 65.5% 85.1% 85.5% 83.9% 96.7% 82.8% 76.4% reviewed/approved 7474 gpt-5.5
2326 Καλλίαρος 47 51 5.501 76.4% 88.2% 96.9% 91.8% 98.7% 78.8% 83.8% reviewed/approved 7458 gpt-5.5
2597 Καρπασία 112 101 -4.678 44.7% 70.4% 69.5% 74.2% 95.2% 76.6% 62.7% reviewed/approved 7783 gpt-5.5
7255 Κυτέριον 22 27 4.647 45.0% 69.3% 65.3% 69.4% 94.7% 74.9% 72.0% reviewed/approved 7820 gpt-5.5
2079 Καισάρεια 39 34 -4.092 51.0% 73.6% 85.5% 90.4% 96.6% 79.0% 77.0% reviewed/approved 7774 gpt-5.5
3530 Κριώα 26 22 -4.056 34.7% 60.7% 47.0% 62.5% 95.2% 69.7% 69.4% reviewed/approved 7806 gpt-5.5
2114 Κάλβιος 18 15 -3.649 36.3% 58.4% 66.5% 66.7% 93.4% 73.9% 68.9% reviewed/approved 7777 gpt-5.5
2624 Κατάβαθμος 20 17 -3.501 55.4% 82.1% 76.1% 86.5% 96.9% 83.5% 83.1% reviewed/approved 7795 gpt-5.5
7248 Κύρου πόλις 32 35 3.389 29.9% 65.8% 69.7% 68.7% 95.3% 78.5% 79.7% reviewed/approved 7813 gpt-5.5
7253 Κυρτώνιος 18 22 3.351 43.4% 66.6% 73.1% 70.0% 96.5% 76.0% 81.2% reviewed/approved 7818 gpt-5.5
7247 Κύρνος 50 45 -3.276 34.0% 59.4% 63.1% 67.4% 95.6% 76.7% 77.8% reviewed/approved 7812 gpt-5.5
7267 Κωνώπη 64 58 -3.238 55.7% 73.4% 70.4% 78.7% 96.4% 76.3% 73.4% reviewed/approved 7832 gpt-5.5
3288 Κορώνεια 110 107 3.174 35.3% 64.6% 62.4% 67.9% 94.1% 75.8% 63.5% reviewed/approved 7803 gpt-5.5
2055 Καβασσός 109 106 3.100 40.3% 62.8% 67.6% 69.8% 94.9% 77.8% 57.2% reviewed/approved 7760 gpt-5.5
2626 Κατάνη 95 93 3.061 40.9% 62.0% 61.3% 66.0% 93.6% 73.2% 61.3% reviewed/approved 7798 gpt-5.5
7259 Κύφος 54 55 3.021 48.7% 74.6% 89.9% 82.6% 95.4% 76.9% 76.8% reviewed/approved 7824 gpt-5.5
2115 Καλὴ ἀκτή 64 64 2.762 100.0% 100.0% 100.0% 100.0% 100.0% 83.1% 97.1% reviewed/approved 7454 gpt-5.5
3496 Κοτιάειον 71 65 -2.719 33.8% 54.8% 57.5% 57.4% 93.1% 73.7% 63.0% reviewed/approved 7804 gpt-5.5

Residual Predictors

Positive signed-residual terms are associated with AI translations longer than expected from the human length; negative signed-residual terms are associated with shorter AI translations. Absolute-residual models identify terms associated with larger AI-human length divergence.

Greek source terms predicting signed length residuals

Features: 176; ridge alpha: 24.245; cross-validated R^2: -0.168.

Positive termCoefficientDocsMean presentMean absent
και 0.4172 66 0.295 -0.573
οι 0.3048 11 3.648 -0.451
απο 0.2951 27 1.644 -0.608
τη 0.2187 15 1.632 -0.288
τε 0.2179 5 2.874 -0.151
δε και 0.2068 8 1.354 -0.118
εν τη 0.2005 7 3.300 -0.248
και το 0.1960 9 3.224 -0.319
εκαλειτο 0.1892 7 1.471 -0.111
εστι δε και 0.1881 3 6.357 -0.197
και πολις 0.1840 5 3.398 -0.179
πολις 0.1839 68 0.461 -0.980
ορος 0.1661 6 5.511 -0.352
ου 0.1654 14 1.401 -0.228
εκαλειτο δε 0.1633 3 2.910 -0.090
αφ ου 0.1584 5 1.929 -0.102
ος 0.1544 6 2.678 -0.171
εστι δε 0.1530 5 2.631 -0.138
πολιχνιον 0.1512 3 5.589 -0.173
πολις και 0.1492 6 4.490 -0.287
Negative termCoefficientDocsMean presentMean absent
το -0.2460 70 -0.026 0.060
οικητωρ -0.2070 6 -4.965 0.317
περι -0.2040 11 -2.843 0.351
εθνος -0.1893 8 -4.381 0.381
δια -0.1724 16 0.551 -0.105
το εθνικον -0.1678 56 -0.173 0.220
εθνικον -0.1678 56 -0.173 0.220
την -0.1582 13 -0.835 0.125
το θηλυκον -0.1568 6 -4.568 0.292
γαρ -0.1353 9 0.204 -0.020
θεοπομπος -0.1340 5 -2.069 0.109
ηρωδιανος -0.1324 3 -9.662 0.299
εκαταιος ασια -0.1303 5 -1.360 0.072
ασια -0.1303 5 -1.360 0.072
θρακης -0.1244 5 -1.567 0.082
πολις θρακης -0.1229 4 -1.825 0.076
μεν -0.1204 4 -8.176 0.341
εστι και -0.1041 20 -0.005 0.001
φασι -0.1023 8 0.434 -0.038
ης -0.1014 3 -2.942 0.091

Greek source terms predicting large absolute residuals

Features: 176; ridge alpha: 8.377; cross-validated R^2: -0.326.

Positive termCoefficientDocsMean presentMean absent
και 1.0146 66 3.388 1.542
δε 0.8619 46 3.869 1.817
δια 0.8497 16 6.204 2.105
τω 0.7200 16 5.411 2.256
απο 0.7008 27 5.154 1.875
εν 0.6631 37 3.787 2.157
εκαλειτο δε 0.6575 3 16.882 2.324
εστι 0.6257 24 4.712 2.144
τη 0.6041 15 5.632 2.254
της 0.5483 35 4.319 1.921
την 0.5452 13 6.485 2.204
κατα 0.5395 10 7.050 2.284
απο της 0.5324 6 9.629 2.322
γαρ 0.5293 9 6.362 2.404
δια του 0.5136 10 6.599 2.334
δε και 0.4959 8 7.617 2.338
εκαλειτο δε και 0.4737 3 16.882 2.324
του 0.4566 37 4.218 1.905
αφ ου 0.4534 5 7.271 2.523
εν τω 0.4497 6 8.480 2.395
Negative termCoefficientDocsMean presentMean absent
το εθνικον -0.5848 56 2.824 2.680
εθνικον -0.5848 56 2.824 2.680
δευτερω -0.3665 3 0.834 2.820
μια -0.2517 3 0.686 2.825
οτι -0.2501 5 2.027 2.799
πλησιον -0.2375 8 2.122 2.816
κωμη -0.2276 3 1.281 2.806
ως -0.2184 52 3.120 2.371
τινες -0.2102 6 1.685 2.829
δε οτι -0.1993 3 0.798 2.821
φησι το -0.1945 3 0.513 2.830
πολις -0.1923 68 3.126 1.983
πορρω -0.1784 3 0.994 2.815
ου πορρω -0.1784 3 0.994 2.815
προς τη -0.1708 4 2.074 2.789
τεταρτω -0.1675 3 1.749 2.792
πολιτης -0.1571 18 3.273 2.648
τους -0.1553 4 1.944 2.794
του ποντου -0.1497 3 1.378 2.803
ποντου -0.1497 3 1.378 2.803

AI English terms predicting large absolute residuals

Features: 341; ridge alpha: 8.377; cross-validated R^2: -0.394.

Positive termCoefficientDocsMean presentMean absent
the 0.8404 93 2.836 1.760
with 0.8357 17 5.729 2.152
and 0.7713 55 3.687 1.628
there 0.6330 34 4.302 1.966
after 0.5796 23 5.414 1.968
usage 0.5716 4 9.291 2.488
there is 0.5530 26 4.780 2.051
by 0.5328 16 6.080 2.128
there is also 0.5145 24 5.055 2.036
of 0.4911 64 3.453 1.529
also 0.4851 43 3.774 1.996
to 0.4647 28 4.806 1.965
is also 0.4639 32 4.334 2.020
on 0.4460 13 5.307 2.380
is 0.4264 89 2.839 2.124
zeus 0.3978 5 11.520 2.299
diphthong 0.3890 3 10.795 2.512
used 0.3871 8 6.965 2.395
for 0.3698 12 5.866 2.337
with the 0.3672 5 7.554 2.508
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.3772 17 1.834 2.950
ethnonym is -0.3772 17 1.834 2.950
the ethnonym -0.3520 18 1.918 2.945
ethnonym -0.3520 18 1.918 2.945
the ethnic is -0.3501 28 2.263 2.954
ethnic is -0.3058 29 2.484 2.873
the ethnic -0.2948 32 2.654 2.810
an -0.2659 19 2.741 2.765
ethnic -0.2539 33 2.837 2.723
near -0.2393 13 2.136 2.854
of his -0.2115 17 2.233 2.868
some -0.1957 10 1.668 2.882
as in -0.1821 20 3.221 2.645
city the -0.1705 3 0.996 2.815
a city in -0.1695 41 3.068 2.547
some say -0.1652 6 1.409 2.847
book of his -0.1637 15 2.405 2.823
it has -0.1634 4 1.262 2.823
a city -0.1454 63 3.214 1.988
city in the -0.1421 3 1.402 2.802

Human English terms predicting large absolute residuals

Features: 335; ridge alpha: 11.938; cross-validated R^2: -0.393.

Positive termCoefficientDocsMean presentMean absent
the 0.7632 90 2.854 1.920
with 0.6448 16 6.023 2.139
and 0.6175 52 3.782 1.654
also 0.4925 43 3.850 1.939
is with 0.4359 4 11.763 2.385
for 0.4046 11 6.882 2.251
there 0.3634 34 3.873 2.187
is also 0.3563 31 4.474 1.991
on 0.3546 16 4.783 2.375
there is 0.3463 24 4.456 2.225
of 0.3422 66 3.330 1.655
says 0.3088 17 5.445 2.211
there is also 0.3009 22 4.725 2.206
after 0.2927 22 5.110 2.098
with the 0.2892 5 8.785 2.443
his on 0.2815 3 11.751 2.482
is 0.2728 87 2.874 2.002
when 0.2727 7 7.165 2.429
their 0.2688 3 16.882 2.324
usage 0.2585 4 7.744 2.553
Negative termCoefficientDocsMean presentMean absent
the ethnonym is -0.3577 52 2.576 2.961
ethnonym is -0.3577 52 2.576 2.961
an -0.2844 22 2.861 2.732
the ethnonym -0.2831 56 2.924 2.553
ethnonym -0.2807 57 2.927 2.540
as in -0.1875 29 3.048 2.643
some -0.1465 10 1.571 2.893
a city in -0.1438 40 3.232 2.446
a city -0.1379 64 3.183 2.010
in book of -0.1329 14 2.379 2.822
ethnonyms -0.1213 3 1.378 2.803
book -0.1192 39 3.233 2.459
one of -0.1176 5 1.494 2.827
one of the -0.1161 4 1.280 2.822
it is -0.1129 24 3.537 2.515
herodotos book -0.1120 5 2.406 2.779
city -0.1116 70 3.107 1.951
city in -0.1109 41 3.352 2.349
which -0.1068 20 2.925 2.719
some say that -0.0996 4 1.056 2.831

Prompt Text

Show prompt text
You are an expert classical philologist and translator specialising in Stephanos of Byzantium's Ethnika.

Goal
- Produce a clear, scholarly English translation in the established reviewed house style.
- Stay close to the Greek while making Stephanos' compressed reference-work prose intelligible in English.
- Translate the Greek main text of the entry, not the modern apparatus, editorial locators, or later bibliographic explanation.

Priority order
- Follow the source Greek first.
- Apply the output and formatting rules below.
- Apply any entry-specific Additional translation context supplied with the entry, but only where the cited Greek evidence is relevant.
- Do not force a rule, gloss, or proper-noun convention when the local syntax or context contradicts it.

Output rules (required)
- Respond ONLY by calling the submit_translation tool with a single string field: {"translation": "..."}.
- The translation text must contain only the translation.
- Do not include analysis, commentary, alternative translations, uncertainty notes, or Markdown outside the translation itself.

A) Formatting + spelling
- Use Australasian spelling and punctuation conventions.
- Preserve paragraphing and line-breaks when they are present in the Greek source.
- For verse quotations, preserve line segmentation in English where the Greek source marks or implies individual verse lines.
- For lengthy prose quotations that function as a quoted block, allow paragraph/quotation separation so the quotation remains readable. Do not create stanza-like line breaks for ordinary short prose snippets.
- Use single quotes for quoted forms, cited snippets, translated quotations, appellations derived from or pertaining to the headword (such as ethnonyms and similar gentilics, but not divine epithets), and Greek forms rendered in Latin letters: '...'.
- Avoid double quotes.
- Use *italics* for titles of ancient works when they are actual work titles: *Cypriaka*, *Bassarika*, *Europe*, *Periegesis*, *Geography*.
- Do not italicise descriptive phrases that are not work titles, such as an epitome of another author's books.

B) Opening / structure
- Let the Greek structure control the opening. Do not force every entry into a single template.
- When the entry opens with the headword and a brief identification, normally begin with the headword transliterated into Latin letters without Greek diacritics, followed by the identification.
- Typical form where appropriate: Headword: ...
- Appositive openings like 'Karia, the country.' are acceptable when the entry is of that type.
- If the headword line gives both nominative and genitive, render it compactly as 'X, Y: ...' unless a grammatical tag is needed for clarity.
- Keep numbered or lettered sequences of homonymous places as inline numbered items like (2) ... (3) ..., matching the Greek's structure.
- When Greek ἔστι καὶ ἄλλη / ἕτερος introduces another place with the same name, use 'There is also another ...' when that is needed for sense.
- For appositional settlement descriptions of the type X (proper noun) + Y (settlement noun) + Z (genitive territory), translate the territory as location: 'X, a city/town/village in Z', not 'of Z', unless the Greek clearly expresses possession, origin, descent, or another non-locative relationship.

C) Transliteration + naming
- Do NOT use macrons or acute accents in ordinary transliteration: Karystos, not Kárystos.
- Prefer Greek-form transliteration when Stephanos is discussing Greek forms, derivation, spelling, ethnonyms, or other metalinguistic material.
- Use these general correspondences unless an established house-style exception is supplied: η = e, κ = k, φ = ph, ῥ = rh, ῤ = r, υ = y, χ = ch.
- Use Greek-style transliteration of diphthongs, especially ai for αι. Do not drift into Latinised forms such as ae or e where the Greek form matters.
- Use forms such as Karchedon, Chalkedon, Kapetolion, Kyrene, Kyrtos, Kappadokia, and Palaistine in Greek-form contexts.
- Avoid Latinised forms such as Carthage, Chalcedon, Capitolium, Cyrene, Cyrtos, Cappadocia, and Palestine in Greek-form contexts.
- Use established conventional English forms only when they are house-style exceptions or clearly ordinary geography, for example Rome, Cyprus, Egypt, Syria, India, Sicily, Thebes, Peloponnese, the Black Sea, the Red Sea, and the Persian Gulf.
- Treat conventional English-name exceptions as a conservative whitelist. Do not expand it by analogy without entry-specific guidance.
- Keep Greek-form transliteration for most ethnonyms and derived adjectives, and put the cited form in single quotes.
- Do not Latinise river, people, or place names merely because a familiar Latin or English form exists.

D) Citations (authors/works/books)
- Translate only the citation information that is present in the original Greek words of the lemma entry.
- Ignore modern/editorial locator codes and apparatus-like add-ons: omit RE/SH numbers, FGrHist numbers, fragment numbers, chapter/section locators, column references, and parenthetical references such as (12,1,2 [C 533,12]), (Op. 344), or (A. R. 2,403).
- Do not infer a book number from a modern parenthetical reference.
- If the Greek says only 'Strabo' plus a modern parenthetical locator, translate 'Strabo', not 'Strabo, book 12'.
- Use 'as per X' only where the Greek itself has ὡς + author/name or an equivalent example-marker construction that supports that rendering.
- If the Greek explicitly gives a Greek book numeral, translate it: 'Ktesias in book 3 of his *Persika*'.
- Keep citations compact, mirroring the source's incompleteness:
  - Author + work title: 'Hellanikos in his *Cypriaka*'
  - Author + book: 'Pausanias, book 9.'
  - Author + work + book: 'Dionysios in book 3 of his *Bassarika*: ...'
- Work-title guidance:
  - Γεωγραφούμενα / γεωγραφικά in citation contexts -> *Geography*, not *Geographoumena*, when this is the established house treatment for the cited work.
  - Περιήγησις / Περίοδος should normally preserve the transmitted title as *Periegesis* when that is what Stephanos gives.
  - If the Greek says an author is cited in an epitome of the eleven books, do not italicise *Epitome* as a work title; translate 'in an epitome of his eleven books'.
- If a citation is present in the source as bracketed main text and later syntax depends on it, translate it in brackets rather than omitting it.
- Do not add external citation labels like Iliad, Works and Days, or Argonautica unless the Greek itself names them.

E) Entry-specific guidance context
- Additional translation context may be supplied for the current entry.
- This context can include matched formula rules, glossary rules, proper-noun conventions, contextual vocabulary bias, and external source passages.
- Apply required or replacement guidance strongly when the evidence matches the current Greek.
- Apply formula and gloss guidance according to the local syntax; do not make it mechanical if the context requires a different construction.
- Apply proper-noun guidance for consistency where supplied, but do not overwrite the Greek's grammatical point.
- Apply contextual vocabulary bias as a preference, not as a one-to-one replacement. It should steer the wording when the stated semantic context is present.
- Use external source passages only for quoted or allusive material. Do not replace Stephanos' own wording with the external source translation.
- Do not mention which guidance rules fired in the translation output. The review interface should surface that separately.

F) Philological, grammatical, and orthographic discussion
- When the entry discusses spelling, letters, diphthongs, accents, stress, case, gender, number, dialect, or morphology, preserve the technical claim precisely.
- Preserve Greek letters or digraphs in Greek script when the letter or digraph itself is under discussion, for example ι, ει, οι, υ.
- Prefer transliteration for cited word-forms unless the point depends on Greek spelling, accentuation, a specific letter, or a visible dialectal/prosodic distinction.
- If accentuation or prosody is the topic:
  - τόνος -> stress or accentuation, according to context.
  - ὀξύς describing a word's τόνος -> oxytone, not merely 'acute'.
  - βαρύς describing a word's τόνος -> barytone, not merely 'grave'.
  - Keep the form in a representation that makes the contrast visible. Use Greek script where the Greek spelling or accent is the point.
- When Stephanos cites an inflected form, dialectal form, or poetic form as evidence for another form, do not normalise away the form being demonstrated.
- If the cited form is inflected or dialectal, translate using the English lemma derived from the appropriate Greek nominative where that is needed for consistency. For example, treat Κυταιέος as evidence for Kytaieus from Κυταιεύς, not as a different nominative such as Kytaios.
- Use brief grammatical tags only when they are necessary to make the philological point visible:
  - case: (nom.), (acc.), (gen.), (dat.), (voc.)
  - gender: (m.), (f.), (n.)
  - number: (sing.), (pl.)

G) Comparisons / derivational analogies
- For ὡς + X where X is an analogous form or etymologically related noun, use a form like "(as in 'X')" where this makes the analogy clear.
- For fuller derivational analogies, keep the relation explicit:
  - as 'Byzantios' is from the name Byzantion
  - as in 'Lykaones', from Lykaonia
- Avoid bare 'like X' when the point is morphological analogy.

H) ἀπό, ἀφ' οὗ, and naming/derivation formulae
- If ἀπό + genitive gives an eponym or naming source, prefer 'after X' or 'named after X' rather than a flat 'from X'.
- In entries with temporal naming contrast, ἐκαλεῖτο X ... κέκληται Y -> 'It used to be called X; it is now called/named Y.'
- For ἀφ' οὗ:
  - With a masculine personal antecedent in a naming context, translate 'after whom'.
  - With an explicit neuter antecedent, translate 'from which'.
  - With a dropped neuter antecedent or Stephanos' example-marker usage, translate naturally according to the syntax; use 'as per' only when the Greek construction supports an evidentiary/example sense.
- Avoid archaic 'whence' unless it is clearly the most natural scholarly English.

I) Syntax and restraint
- Translate directly; do not add background explanation, modern bibliography, or parenthetical glosses unless Stephanos gives them.
- Do not add glosses like (synoikia), (kalathos), or explanatory etymologies unless the Greek explicitly defines the term.
- Avoid gratuitous adversatives for δέ.
- Use 'however' only for a real contrast, such as a correction, opposition, exception, or contrastive second item that English needs to mark.
- Avoid 'on the other hand' unless the Greek contrast is strong and English needs it.
- Convert participles into natural English when strict literalness obscures the structure.
- Circumstantial participles may need 'because', 'as', 'when', or a relative clause.
- Preserve uncertainty when the Greek is uncertain, corrupt, or genuinely ambiguous; do not invent a resolution.
- If the source includes bracketed or braced text as part of the lemma main Greek text of the Meineke edition rather than apparatus, translate it and keep brackets if that best reflects the source.
- If the source gives indirect speech under φασί, do not automatically turn it into a formal quotation unless the source punctuation or reference-work style calls for it.

Now translate the provided entry accordingly, and submit it via submit_translation.